A Multi-Agent Self-Adaptive Genetic Algorithm for Multi-Objective Optimization

Now a days multi-objective optimization is one of the biggest problem, to solve such type of problem, the genetic algorithms and an agent technology are integrated and is applied to solve such type of multi-objective optimization problems. In this algorithm an agent represents an applicant answer to the multi-objective optimization problem. An agent lives in the grid environment and it owns confined space called the neighborhood. An agent can compete and cooperate with other agents with other agents, to achieve the purpose of inheritable factor replaced and developed. Agents also retains some knowledge of the environment and can learn by itself while developing, in order to adapt itself to the environment better and enhance its possibility. A new multi-objective genetic algorithm based on Multi-Agent Self-Adaptive Genetic Algorithm (MASAGA) is proposed, in this algorithm evolution parameters are accustomed adaptively in the evolutionary process and a new selection operator is used to select individual. By accustoming the mutation and crossover parameters in the evolutionary process, it can improve the precision and convergence speed of the algorithm. Several benchmark functions are run to test the execution of the algorithm, the simulation results indicate that the multi-objective evolutionary algorithm based on MASAGA has better performance. The algorithm can converge to the Pareto solutions quickly, and some intelligent algorithms are embedded in NSGA-II to improve the algorithm in order to expect better results.